[Mike Copioli]

Originally Posted by Gregor

http://www.chiefdelphi.com/forums/sh...ad.php?t=83973

Oh that problem. That was the PTC. We have used Jags on our window motors since 2009 and have never seen a problem like this that was not caused by the PTC. Also I would like to correct some factual errors in that post.

The poster describes the output of the Jag to be switching between +12 and -12 volts. This type of rectification is known as locked anti-phase. Both the Victor and the Jaguar use a type of rectification called sign-magnitude. Although both motor controllers use the same type of rectification the Jaguar Does whats known as synchronous sign-magnitude rectification.

During the off period of the duty cycle the Victor allows the free wheeling current(opposite of the forward current) of the motor to return through the body diode of the FET's. The diode has a forward voltage drop of about 1 volt regardless of the current passing through it. Since Power = I x V if 20 amps of current is passing through the body diode, then the power being dissipated in the FET is 20 watts (1volt x 20 Amps). This causes heat rise and losses to accumulate in the Victor.

The Jaguar does something a little different, during the off period the jaguar switches the opposing low side FET on thus connecting the motor leads together, the freewheeling current is now passing through the drain source junction of the FET's rather than the body diode. Since the FET's have a very low drain source resistance (say 2 mOhms) the power dissipated by the FET's is = to (I*I)*r the power dissipated is .8 watts. What does this mean? Well basically more power is being dissipated in the motor windings instead of the FET's. This is probably what is causing the PTC to over heat and trip. However what it really means is that the window motor is being over driven. Moving it to a Victor allows a slower decay of the free wheeling current thus decreasing the heat dissipated by the motor.

So the problem has nothing to do with the switching frequency and more to do with the type of rectification. I would suspect in this example the Victor was actually providing less power than the Jag possibly due to calibration differences. Both devices use different PWM input values so it is possible that the Victor may not have been delivering full power. Another thing to remember is that at full power both devices are full on so switching has no affect. This could also be a contributing factor since the Victor is not linear, the same PWM value will yield a very different output.